INDEX

Explanations

mentions of legal, medical, or administrative terms

oai_token-act-pair · gpt-3.5-turbo

terms related to legal and administrative language

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

LET

-0.69

netflix

-0.68

itans

-0.67

Â¯

-0.67

PLIED

-0.66

cmp

-0.65

UTF

-0.65

udeb

-0.64

-0.63

SpaceEngineers

-0.62

POSITIVE LOGITS

 Advice

1.01

 Problems

0.97

 Profile

0.96

 Benefits

0.96

 Releases

0.95

 Issues

0.94

 Analysis

0.94

 Overview

0.94

 Features

0.94

 Description

0.94

Activations Density 0.362%

mentions of legal, medical, or administrative terms

terms related to legal and administrative language

No Comments

No Known Activations